LFM2.5-VL-3B for Better and Faster Vision Capabilities for the Edge
- ID
- 13475
- Status
- summarized
- Published
- 12 Aug 2026, 10:00 PM
- Fetched
- 12 Aug 2026, 10:36 PM
- Provider
- Hugging Face Blog
- Category
- developer-ai
- Original URL
- https://huggingface.co/blog/LiquidAI/lfm2-5-vl-3b
- Source URL
- https://huggingface.co/blog/feed.xml
Summary
- Score
- 5.5
- Created
- 12 Aug 2026, 10:36 PM
- Tags
- Audience
- developersai_ml_learnersai_agent_users
What happened
LiquidAI released LFM2.5-VL-3B, a 3.1B parameter vision-language model designed for on-device/edge use, pairing a SigLIP2 400M vision encoder with their LFM2.5-2.6B text backbone. It was pre-trained on ~34T tokens with 4x more vision data than prior versions, supports 128K vocabulary for non-Latin scripts, and adds screen/UI understanding, object grounding, multi-image input, and function calling. Benchmarks show it leading its size class on real-world image tasks (RealWorldQA 73.1, MMStar 63.3) against comparably-sized models from Qwen, InternVL, and Gemma.
Why it matters
If you are building on-device apps that need document/screen understanding or vision-grounded function calling without cloud API latency or cost, this is a concrete 3B model worth benchmarking against Qwen3.5-2B or InternVL 3.5 2B for your use case. The function-calling capability in vision-text contexts is the differentiator to test, since most small VLMs struggle there.
Discussion angle
Compare LFM2.5-VL-3B's function-calling + screen understanding against Qwen3.5-2B for a practical agent use case like reading a UI screenshot and deciding which tool to call — does the 3B size class actually work for real-time on-device agents, or is it still too weak?